Using data from a longitudinal study of inmates in the D.C. metropolitan area, the proposed project aims to investigate the relationship between jail inmates' perceived risk of contracting HIV and their actual risk behavior, in the first year post-release. The study also seeks to better understand the role of connectedness to both the community-at-large and to the criminal community in newly-released inmates' engagement in HIV risk behavior (i.e., risky sex and drug use). Further, the proposed research will explore the impact of degree of community social disorder on HIV risk behavior, along with its interactive effects with community connectedness to predict such behavior. This project will yield new information regarding individual-level cognitive and community attachment factors as well as structural and social influences that may influence HIV risk behavior, thus potentially providing multiple novel targets of intervention in order to reduce the rate of HIV infection in this high risk population.